×

Document transcription system training

  • US 8,335,688 B2
  • Filed: 08/20/2004
  • Issued: 12/18/2012
  • Est. Priority Date: 08/20/2004
  • Status: Active Grant
First Claim
Patent Images

1. In a system including a first document, the document tangibly stored in a computer-readable medium and containing at least some information in common with a spoken audio stream, a method performed by a computer processor executing instructions tangibly stored in a first computer-readable medium, the method comprising steps of:

  • (A) identifying text tangibly stored in the first document on a second computer-readable medium, wherein the text represents a concept;

    (B) identifying, based on the identified text, a plurality of at least three spoken forms of the concept, including at least one spoken form not contained in the first document, wherein all of the plurality of spoken forms have the same content as each other, wherein (B) comprises;

    (B) (1) identifying a name of the identified text; and

    (B) (2) using the identified name to identify a corresponding context-free grammar in a grammar repository, wherein the corresponding context-free grammar specifies the plurality of spoken forms of the concept;

    (C) replacing the identified text with the corresponding context-free grammar to produce a second document tangibly stored in a third computer-readable medium; and

    (D) generating a first language model, tangibly stored in a fourth computer-readable medium, based on the second document.

View all claims
  • 12 Assignments
Timeline View
Assignment View
    ×
    ×